nt... Dn. lliol. 39: 575-579 (1995) 575 O,igil/al Arlirlr efgf is expressed in the dorsal midline of Xenopus laevis HARRY v. SAACS'. MARY ELZABETH POWNALL and JONATHAN M.W. SLACK mperial Cancer Research Fund Developmental Biology Unit, Department of Zoology, University of Oxford. Oxford. United Kingdom ABSTRACT A detailed study of the expression pattern of embryonic fibroblast growth factor (efgf)during early Xenopus development has been undertaken using whole-mount DG in situ hybridization. We show that the zygotic expression of efgfis activated in the mesoderm of the early gastrula and is first visualized as a ring around the blastopore, with significantly higher levels of expression on the dorsal side of the embryo. As gastrulation proceeds, efgf transcripts become increasingly abundant in the dorsal blastopore lip. n the early neurula efgf expression can be detected in the extreme posterior of the embryo around the closed blastopore and in the cells of the notochord. This latter result is significant and represents the first report of a Xenopus FGF that is expressed in the notochord. n addition, we show that during gastrula and neurula stages. expression of efgf closely follows the expression of the Xenopus brachyury (Xbra) gene. During later development efgf expression is localized to the tail-bud region and a stripe at the mid-brain/hind-brain junction. These data provide further evidence that FGFs play an important role in regulating the expression of brachyury in the developing mesoderm. KEYWORDS: hmc/i.1ulj. ('FGF,fibroblast Kmwth f(l(tm; lotorhord, Xt'1lOjms ntroduction n recent years it has become clear that members of the fibroblast growth factor (FGF) family are important for the development of organisms as diverse as Drosophila and the mouse (Basilico and Moscatelli. 1992; Shishido etal.. 1993). n the vertebrates there is good evidence that the FGFs are involved in a wide range of developmental processes. For example, recent work indicates that the FGFs play an essential role in the specification and subsequent outgrowth of the vertebrate limb (Niswander et al., 1994;Cohn et al.. 1995).n addition. work carried out on the amphibian Xenopus laevis has demonstrated that the FGFs play important roles during the very earliest stages of development, when the basic vertebrate body plan is established (for review see. Slack. 1994). n this laboratorywe have made an extensive study of the expression patterns and biological activities of the FGFs in Xenopus (saacs et al.. 1992; Tannahillel al.. 1992; Song and Slack. 1994; reviewed, Slack, 1994). These data have allowed us to propose that a major role of the FGFs is the regulation of gene expressionwithinthe newly formed mesoderm duringgastrula and neurula stages (saacs et al., 1994). n particular there is good evidence that the FGFs are important for regulating the expression from the Xenopus homologue of the brachyury gene (Xbra) (Smith et al.. 1991). A number of pieces of data point to a close link between FGF activity and the regulation of brachyury expression. First, it is clear that a functional FGF signal transduction pathway is required for the activation and maintenance of Xbra expression in the Xenopus embryo (Amaya et al.. 1993; saacs el al., 1994). Second. there is a striking similarity between the phenotype of Xenopus embryos in which FGF function is inhibited and the phenotype of zebra fish and mouse embryos homozygous for mutations in the brachyury gene. Both phenotypes are characterized by rather normal anterior development and dramatic reductions in posterior structures (Amaya et al., 1991; Herrmann and Kispert, 1994; saacs et al.. 1994; Schulte- Merker et al.. 1994). Third. there is considerable overlap between the expression domains of FGFs and Xbra (Smith et al.. 1991; saacs et al.. 1992; Tannahill et al.. 1992). These earlier studies showed that during late gastrula stages both efgf and Xbra are expressed in the mesoderm of the blastopore region and in later development are expressed in the developing tailbud. Finally it has been shown that FGF signalling is not only necessary but is also sufficient to maintain Xbra expression in cells of the blastopore region (saacs et al.. 1994; Schulte- Merker and Smith, 1995). n addition to being expressed in the mesoderm around the blastopore, Xbra transcripts are found in the notochord cells of the dorsal midline during late gastrula and neurula stages. The previous study of efgf expression using radioactive in situ hybridization to embryo sections however. did not detect efgf transcripts in the involuted cells of the dorsal midline (saacs el al.. 1992). This suggested that efgf was not involved with the regulation and maintenance of Xbra expression in the notochord..h1jrl,t';nlimubm in hi! Pnlwr: FGF. tihrobta't growth factor; ('F(;r. el11hr~ollie lihroblasl gro\\"lh rat"lor..address for reprints: mperial Cancer Research Fund Developmental Biology Unit, Department of Zoology, University of Oxford, South Parks Road, Oxford OXl 3PS, United Kingdom. FAX: 01865 281310" O~14.6~82195S03.00 C LHC Pr~,\ Prinltdin S";n --- --
576 H. V saacs et al. B r ḋl 1 A lib enh pw..... ~ }. nee mhj, J A ped L
efgf e.\pression in Xenopus 577 We have now extended our earlier characterization of the efgf expression pattern using the sensitive technique of wholemount DG in situ hybridization and confirm that in the eariy gastrula efgf and Xbra are coexpressed in the blastopore region. Later expression of efgf is also detected in specific regions of the developing central nervous system and tailbud. However, the whole-mount technique unexpectedly reveals that efgf expression becomes increasingly prominent within the dorsal mesoderm during mid and late gastrula stages and continues to be expressed in the notochord through neurula stages. These data raise the possibility that efgf not only regulates Xbra expression in the blastopore region but also in the notochord and suggests that efgf is likely to have a role in the specification and patterning of dorsal midline structures. Results efgf is expressed in the mesoderm of the early gastrula Low levels of maternal efgf mrna can be detected by RNAase protection in fertilized eggs (saacs et af.. 1992). However, using the techniques of radioactive and DG in situ hybridization, we have not been able to visualize the spatial distribution of maternal efgf mrna during early cleavage or blastula stages. We are first able to detect efgf expression using whole-mount DG in situ hybridization, following the onset of zygotic transcription in the early gastrula. At stage 10.5 efgf mrna is present in a complete ring around the blastopore, although the signal is considerably stronger on the dorsal side of the embryo (Fig. A,B and C). This pattern is very similar to the gradient of Xbra expression in the marginal zone of the early gastrula reported by Ruiz i Altaba and Jessell (1992) and supports the view that there is a close link between the regulation efgf and Xbra expression. efgf is expressed in the blastopore region and the notochord of the fate gastrula and neurula Figure 10 shows that by stage 11.5 Xbra is expressed in a complete ring around the closing blastopore and in addition is expressed in notochord cells of the dorsal midline. Our previous study of efgf expression failed to detect any efgf expression in the notochord suggesting that other factors are involved in the regulation of Xbra expression in the dorsal midline. However, Figure 1E shows that efgf is expressed at high levels in the involuting dorsal mesoderm of the stage 11.5 gastrula. This rais- es the possibility that efgfis also required for the activation and maintenance of Xbra expression in the notochord. At this stage the expression of efgf is very faint in the lateral and ventral aspects of the blastopore lip. By the end of gastrulation efgf transcripts are clearly detected in the cells of the notochord and in wings of tissue that extend laterally and ventrally around the blastopore (Fig. 1F). n the open neural plate stage 13 it can be seen that Xbra (Fig. 1G) and efgf (Fig. 1H) are both expressed in very similar domains in the posterior of the embryo around the closed blastopore and in the notochord. efgf is expressed in the posterior notochord of the early tailbud stage During later neurula stages and in the early tailbud the expression of efgf becomes increasingly restricted to the extreme posterior of the embryo. The dorsal posterior view of a cleared stage 23 embryo in Figure 11shows efgf mrna is present in the posterior notochord and tailbud region. The side view of an embryo from the same stage shows that efgf expression is absent from all but the most posterior part of the dorsal axis (Fig.1J). efgf is expressed in the CNS and tail forming region during tailbud extension Figure 1K shows that during tailbud stages efgf expression is not detected in the mature notochord but is present in both the chordoneural hinge and the posterior wall of the neuroenteric canal in the tail forming region (Gont et af., 1993; Tucker and Slack, 1995). The expression of efgf within these structures is very much in keeping with lineage studies which show that the chordoneural hinge derives from cells of the late dorsal lip and that the posterior wall of the neuroenteric canal arises from the lateral lips of the late blastopore (Gont et al., 1993). n addition to this posterior domain, a prominent stripe of efgf expression is seen at the midbrain/hind brain junction (Fig. 1L). t is likely that the staining in the forebrain cavity and branchial arches at this stage represents background staining. Discussion Previous studies have shown that a functional FGF signal transduction pathway is necessary for the activation of expression from the Xenopus homologue of the brachyury gene (Xbra) in normaldevelopment(amayaet al., 1993: saacs et al., 1994) Fig. 1. efgf and Xbra expression during development. A,B,C,E,F,H,,J,K and L Embryos hybridized to efgf antisense RNA. D and G Embryos hybridized to Xbra antisense RNA A and B Vegetal views (dorsal is to the top) of stage 10.5 embryos showing efgf expression in the mesoderm of the marginal zone around the blastopore. The arrow indicates the position of the dorsal lip (bp) The specimen in (B) has been cleared. C) shows the cut surface of a stage 10.5 embryo that has been manually dissected along the dorsoventral midline ra reveal staining in the deep layers of the marginal zone (dorsal is to the right and animal ra the rap) showing that there is a higher level of efgf expression in the dorsal marginal zone (dmz) than in the ventral marginal zone (vmz) (note the animal hemisphere has collapsed onto the vegetal yolk mass during processing). (D Dorsovegetal view of Xbra staining n the blastopore region and involuting dorsal mesoderm of a stage 1 7.5 embryo. E) Oorsovegetal view showing high levels of efgf expression in the involuting dorsal mesoderm of a stage 17.5 embryo above the dorsal lip (dl) of the blastopore. (F Posterior view of a stage 13 embryo (dorsal to the top) showing efgf expression in the notochord (nee) and around the blastopore (bp). tg and H) Dorsal views of stage 13 embryos (anterior to the rap) showing expression of Xbra and efgf respectively. in the notochord and posterior around the closed blastopore (bpj. () Posterior dorsal view of a cleared stage 23 embryo (anterior is into the page) showing efgf expression in tail-bud (rib) and in the posterior notochord (nee). {J Side view of a cleared stage 23 embryo (anterior to the right) showing that efgf expression is restricted to the extreme posterior of the developing axis. The arrow shows the postion Df the proctodeum (pcd). (K) Side view of the tail forming region from a cleared stage 31 embryo showing efgf expression in the chordoneural hinge (cnh) and posterior wall (pw) of the neuroenteric canal (nee). (L) Side view of a cleared stage 31 embryo showing a prominent stnpe of efgf expression at the midbrain/hindbrain junction (mhj) (note the staining in the brain and branchial arch cavities is background). ---
-- 578 H. V. {.aaes er al. and in response to mesoderm induction by activin (Cornell and Kimelman, 1994; Labonne and Whitman, 1994). Data from experiments, which involve disaggregated cell culture of explants from the blastopore region of gastrulae, indicate that the maintenance of Xbra expression in the nascent mesoderm is dependent upon intercellular signals (saacs et al., 1994; Schulfe-Merker and Smith, 1995). These experiments also showed that the addition of efgf protein to such disaggregated cell cultures can mimic the endogenous Xbra maintenance signal. Furthermore, efgf shares overlapping domains of expression with Xbra in the blastopore and tail forming regions of the embryo. These data provide good evidence that efgf functions as an autocrine factor involved in the regulation of Xbra expression within the mesoderm of the blastopore region. However, Xbra is also expressed in the notochord. which is the most dorsal mesodermal tissue-type. Using radioactive in situ hybridization it was not possible to detect efgf expression within the notochord, suggesting that the regulation of Xbra within the cells of the notochord was independent of efgf. n this study we use the technique of whole-mount in situ hybridization to confirm that during early gastrula stages efgf is coexpressed with Xbra in the blastopore region. However, the new data show that efgf is also expressed in the notochord cells of the late gastrula and early neurula. This suggests that efgf may be involved in regulating Xbra expression in the mesoderm of the dorsal midline. Later in development the expression of efgf is lost from the mature notochord but con. tinues to be coexpressed with Xbra in the chordoneural hinge and posterior wall of the tailbud. These data show that throughout tailbud stages efgf is expressed in populations of cells that are related by lineage to cells derived from the blastopore lip at the end of gastrulation. t is likely thai efgf plays a role in regulating the properties of these cells necessary for the complex morphogenetic activity of gastrulation and tail bud extension. n this paper we report the first example of a Xenopus FGF family member being expressed in the cells of the notochord. These data add considerable weight to the view that the FGFs, and in particular efgf, are involved in the regulation of Xbra expression, not only within the mesoderm of the blastopore region, but also in the notochord. The presence of efgf within the notochord also has implicafions for the signalling pathways involved in the development of dorsal midline structures. The signalling molecule sonic hedgehog (shh) has also been shown to be expressed in the notochord of Xenopus (Ekker er at., 1995). n this context it is interesting to note that FGF-4, to which efgf is most closely related, and shh have been implicated in the growth and patterning of limb and feather buds (Niswander et a/., 1994; Nohno et al., 1995). Further work will be required to determine if the efgf and shh signalling pathways interact in a similar way during the induction and patterning of dorsal midline structures. Materials and Methods Embryos Embryos were cultured as per the methods of Godsave et al. (1988) and staged according to Nieuwkoop and Faber (1967). n situ hybridization Whole-mount in situ hybridization procedures were carried out on albino embryos according to Harland (1991). Embryos were treated with proteinase K (10 Jg/ml) for 10 to 20 min. Hybridization was carried out overnight at 60cC. n a number of cases the ANAsing step was omitted as this was found to considerably increase the sensitivity of the technique without a significant increase in background. Color reactions for Xbra were carried out for 1 h and overnight for efgf. Embryos were cleared in 100% 1,2,3,4-tetrahydronaphthalene (Aldrich).The antisense efgf DG labeled probe was transcribed from Eco R linearized XeFGF(i) GS (saacs et al., 1992). The Xbra probe was transcribed from Ca linearized Xbra psp73 (Smith et al., 1991). Acknowledgments We are extremely grateful to Abbie Tucker and Patrick Blader for critical reading of this manuscript. References AMAYA, E., MUSC, T.J. and KRSCHNER. MW. (1991). Expression of a dominant negative mutant of the FGF receptor disrupts mesoderm formation in Xenopus embryos. Ce1/56: 257-270 AMAYA, E., STEN, P.A., MUSC, T.J. and KRSCHNER, M.W. (1993). FGF signalling in the early specification of mesoderm in Xenopus. Development 118: 477-487. BASllCO, C. and MOSCATELl, D. (1992). The FGF family of growth-factors and oncogenes. Adv. Cancer. Res. 59: 115-165. COHN, M.J., ZPSUA-BELMONTE, J.C., ABUD. H., HEATH, J.K., and TCKLE, C. (1995). Fibroblast growth.factors induce additional limb development from the flank of chick-embryos. Cel/ao: 739-746. CORNEll, R.A. and KMELMAN, D. (1994). Activin.mediated mesoderm induction requires FGF Deyelopment 12: 453.462. EKKER, S.C., McGREW, L L, LA, C-J., LEE, J.J., VON KESSLER, D.P., MOON, A. T. and BEACHY, P.A. (1995). Distinct expression and shared activities of members of the hedgehog gene family in Xenopus laevis. Deyelopment 121: 2337-2347. GODSAVE, S.F, SAACS. HV. and SLACK, J.M.W. (1988). Mesoderm inducing factors: a small class of molecules. Deyelopment 102: 555-566. GONT, LK., STENBESSER, H., BLUMBERG, B and DE ROBERTS, E. (1993). Tail formation as a continuaton of gastrulation: the multiple cell populations of the Xenopus tailbud derive from the late blastopore lip. Development 119: 991. 1004. HARLAND, A.M. (1991). n situ hybridization: an improved whole-mount method for Xenopus embryos. Methods Celf BiOi. 36: 685-695. HERRMANN, B.G. and KSPERT, A. (1994). The T genes in embryogenesis. Trends. Genet. 10: 280-286. SAACS, H.V., POWNALL, M.E. and SLACK, J.M.W. (1994). efgfregulates Xbra expression during Xenopus gastrulation. EMBO J. 19: 4469-4481. SAACS. H.V., TANNAHLL. D. and SLACK, J.MW. (1992). Expression of a novel FGF n the Xenopus embryo. A new candidate inducing factor for mesoderm formation and anteroposterior specification. Development 114: 711.720. LABONNE, C. and WHTMAN, M. (1994). Mesoderm induction by activin requires FGF-medialed intracellular signals. Development 120: 463-472. NEUWKOOP, P.O. and FABER, J. (1967). Normal Table of Xenopus laeyis (Daudin). North Holland Publishing Company. Amsterdam. NSWANDER, L., JEFFREY, MARTN, G.R. and TCKLE, C. (1994). A positive teedback loop coordinates growth and patteming in the vertebrate limb. Nature 371: 609-612. NOHNO, T., KAWAKAM, Y, OHUCH, H., FUJWARA, A., YOSHOKA, H. and NOJ, S. (1995). nvolvement of the sonic-hedgehog gene in chick feather for. mation. Biachern. Biophys. Res. Commun. 206: 33-39. RUZ ALTABA, A. and JESSELL, T.M. (1992). Pintallavis, a gene expressed in the organizer and the midline cells of frog embryos: involvement in the developing neural axis. Development 116: 81-93. SCHULTE.MERKER, S. and SMTH,J.C. (1995). Mesodenn formation in response to brachyury requires FGF signauing. Curro Bioi. 5: 62.67. SCHULTE-MERKER, S., VANEEOEN, FJ.M.. HALPERN, M.E., KMMEL, C.B., and NUSSLEN-VOLHARD, C. (1994). No tail (ntl) is the zebraflsh homolog of the mouse-t (brachyury) gene. Development 120: 1009-1015.
efgf expression ill Xellopus 579 SHSHDO, E., HGASHJMA, 5., EMOR, Y. and SAGO, K. (1993). Two FGFreceptor homologues of Drosophila: one is expressed in mesodermal primordium in early embryos. Development 117:751-761. SLACK, J.MW. (1994). nducing factors in Xenopus early embryos. CurroBioi. 4: 116-126. SMTH, J.C., PRCE, B.M.J., GREEN, J.B.A., WEGEL, D. and HERRMAN, B.G. (1991). Expression of a Xenopus homolog of Brachyury (T) is an immediate early response to mesoderm induction. Cell 67: 79-87. SONG, J. and SLACK, J.MW. (1994). Spatial and temporal expression of basic fibroblast growth factor (FGF-2) mrna and protein in early Xenopus development. Mech. Dev. 48: 141.151. TANNAHLL, D., SAACS, H.V.. CLOSE. M.J., PETERS, G. and SLACK, J.MW. (1992). Developmental expression of the Xenopus int-2 (FGF-3) gene: activation by mesodermal and neural induction. Development 115: 695-702. TUCKER, A.S. and SLACK, J.M.W. (1995). The Xenopus laevis tail-forming region. Development 121: 249-262. A,ruPt'd fnr publimliun: '\},'11111995 - -- --